Substituted sulfoxides. IV. Phenyl diazomethyl sulfoxide - American

Jan 13, 1975 - (3) B. J. Aylett and J. M. Campbell, Chem. Common., 159 (1967),. (4) For a review of previous work, see R. J. P. Corriu and M. Henner, ...
1 downloads 0 Views 309KB Size
3238 The results of the trapping reactions are consistent with and provide prima facie evidence for the generation of a silicenium ion in methylene chloride a t low temperatures. Previous attempts to chemically trap silicenium ions were not s u ~ c e s s f u l .Work ~ ~ ' ~ is in progress to clarify further the nature of the yellow-green intermediate.

Acknowledgment. The author wishes to thank Professor K . Mislow for helpful discussions. References and Notes (1) This work was supported by National Science Foundation Grant MPS74-18161 to K. Mislow, and the Summer Fellowship Fund of the University of Missouri-St. Louis. (2) J. Y . Corey and R . West, J. Am. Chem. Soc., 85, 4034 (1963). (3) E. J. Aylett and J. M. Campbell, Chern. Commun., 159 (1967). (4) For a review of previous work, see R. J. P. Corriu and M. Henner, J. Organornet. Chem., 74, l(1974). (5) Under electron impact conditions in the gas phase, processes producing silicenium ions are the favored pathways.6 (6) (a) J. Y. Corey, M. Dueber. and M. Malaidza. J. Organomet. Chem., 36, 49 (1972); (b) W. P. Weber, A . K. Willard, and H. G. Boettger, J. Org. Chem., 36, 1620 (1971), and references therein: (c) J. M. Gaidis, P. R . Briggs, and T. W. Shannon. J. Phys. Chem., 75, 974 (1971): (d) H. Sakurai, M. Kira, and T. Sato. J. Organomet. Chem., 42, C24 (1972). (7) J. J. Looker, J. Org. Chem.. 33, 1304 (1968). (8) The pKR+was estimated from the observation that cation I b does not undergo hydride abstraction from cycloheptatriene. The method used to monitor the hydride abstraction reaction was not reported and an incorrect value of the ~ K R for + C7H7+ was quoted.' The pKR+of cation i b should exceed the value of 4 . 7 for C7H7+, but has not yet been determined. (9) J. Y. Corey, Synth. Inorg. Metal-Org. Chem., 2, 85 (1972). (10) Methylene chloride was purified by the method described by Harmon, et al." All reactions were carried out under a N2 atmosphere in oven dried glassware. Diglyme was distilled from lithium aluminum hydride just prior to use. (11) K. Harmon, F. E. Cummings, D. A. Davis, and D. J. Diestler, J. Am. Chem. SOC.,84, 3349 (1962). (12) In the absence of triphenyicarbenium perchlorate, silane 2a does not react with NaBD4 under these experimental conditions. (13) A. G. Brook and K. H. Pannel, Can. J. Chem., 48, 3679 (1970). (14) Address correspondence to the author at: Department of Chemistry, University of Missouri-St. Louis, St. Louis, Missouri 63121

Joyce Y. Corey l 4 Department of Chemistry, Princeton University Princeton, New Jersey 08540 Received January 13. I975

Phenyl Diazomethyl Sulfoxide (Phenylsulfinyldiazomethane)and Phenylsulfinyl Carbene, a Highly Stereoselective Cyclopropanating Agent' Sir:

Although both ~ u l f e n y l ~and , ~ ~ u l f o n y lcarbenes ~~~ are well-characterized intermediates which have been known for some time, the corresponding sulfinyl carbenes are unknown. This deficit arises because the anions of a-halo sulfoxides apparently do not expel halideS and because a-diazo sulfoxides6 and one class of reasonable precursors, a-amino sulfoxide^,^ are unknown. Another class of possible precursors, a-oxo sulfoxides (thiolester S-oxides), a r e known only as unstable reaction intermediates.'O W e wish to report the preparation of the first characterized a-diazo sulfoxide and its facile conversion into phenylsulfinyl carbene, a highly stereoselective cyclopropanating agent. CH2N2

+ PhS(0)CI

+

PhS(O)CH2CI + PhS(O)CHN2 1

Addition of benzenesulfinyl chloride to excess diazomethane in T H F at -78' gave, after removal of solvent, an oily mixture of phenyl chloromethyl sulfoxide (1) and what we believe to be phenyl diazomethyl sulfoxide (phenylsul-

finyldiazomethane) (2) based on the following observations. T h e infrared spectrum of the mixture showed a strong band a t 2040 cm-' which was not in the spectrum of pure 1. Although a-diazo ketones' * and a-diazo sulfones12 absorb near 2100 cm-', this shift from the normal diazoalkane frequency of 2040 cm-' is usually attributed to the acidifying effect of the carbonyl or sulfonyl group, an effect which would not be nearly as strong in the case of the sulfinyl group. The N M R of the mixture shows a band a t 6 5.5 which is not found in 1. This chemical shift is in the region expected from analogy to a-diazo sulfones12 and other diazoalkanes. l 3 R2C=N2

+ HA

+

R2CHA

+ N2

Further evidence for the diazo sulfoxide structure derives from the chemistry of this new compound. One of the characteristic reactions of the diazo group is its facile replacement by strong acids.I4 When the crude mixture of 1 and 2 is treated with excess hydrochloric acid, 1 can be isolated in a quantitative yield based on starting benzenesulfinyl chloride. Similar treatment with hydrobromic acid gives a 64:36 mixture of phenyl bromomethyl sulfoxide and 1 as estimated by N M R , while hydriodic acid reacts to give 65% phenyl iodomethyl ~ u l f o x i d e ' and ~ 35% 1. Control experiments show that 1 is not converted into the iodomethyl compound under the reaction conditions. Thus, the initial product of addition of benzenesulfinyl chloride to CH2N2 in T H F is a 2:l mixture of phenyl diazomethyl sulfoxide (2) and phenyl chloromethyl sulfoxide ( 1 ) . An important and characteristic reaction of diazo compounds is their ability to cyclopropanate olefins.2 Addition of cooled tetrahydrofuran solutions of mixtures of 1 and 2 to room temperature cyclohexene or E - or 2-2-butene a t 0' gave phenylsulfinyl cyclopropanes in 35-40% isolated yields based on starting benzenesulfinyl chloride. Typically, 3540% of 1 was also isolated, further confirming the approximately 2: 1 ratio of 2 to 1 estimated from H X quenching experiments. The stereospecificity of the cyclopropanation was checked with the 2-butenes.I6 E-2-Butene gave cyclopropanes with >97% of the methyl groups trans, and 2 - 2 butene gave cyclopropanes with >99% of the methyl groups cis. The reaction appears to be a singlet carbene cycloaddition. The possibility that the reaction occurs by a two-step mechanism involving the formation and decomposition of a A ' - pyrazoline, while unlikely a t the low temperatures of these reactions, has been considered. One might expect that a pyrazoline (probably L2-3-phenylsulfinyl-pyrazoline) would be isolable under the reaction conditions by analogy to sulf~nylpyrazolines,'~ However, none has been found. TO test the nature of the intermediate further, we have allowed a 1:l mixture of cyclohexene and diethyl fumarate to react with 2. Recently, Houk and coworkers have presented a general perturbation molecular orbital treatment of 1,3 dipolar c y c l o a d d i t i ~ n s . Their ' ~ ~ ~ ~predictions a r e in accord with the experimental observations that substituted styrenes as dipolarophiles with diazomethane show a p = 0.902' and that dimethyl fumarate is 1750 times more reactive than styrene as a dipolarophile with diphenyl diazomethane2' and that, in general, olefins with electron-withdrawing substitutents are superior 1,3-dipolarophiles with diazoalkanes. O n the basis of these observations and calculations, one can confidently predict that diethyl fumarate should be much more reactive with 2 than should be cyclohexene, if the reaction is a 1,3-dipolar addition. When a 1: 1 mixture of diethyl fumarate and cyclohexene is allowed to react with 2, no adducts of diethyl fumarate with either 2 or phenylsulfinyl carbene are found. In fact, although the yield of norcaranes is reduced from 50 to 30% as estimated from N M R , they are definitely still the major products even in

Journal of the American Chemical Society J 97:ll / M a y 28, 1975

+

3239 the presence of the added diethyl fumarate.22 The stereoselectivity of phenylsulfinyl carbene was determined in its addition to both 2-2-butene and cyclohexene. T h e ratio of anti addition to syn addition,23 anti/syn, is 34/1 for addition to cyclohexene and >99/1 for addition to Z-2-butene. Compared to the stereoselectivities of other carbenes, these ratios a r e extremely Hoffman, Levin, and Moss,Zs on the basis of extended Huckel calculations, state that "in highly exothermic association reactions, where the transition state occurs a t large separation of reaction partners, the activation energy for the path leading to the sterically more hindered product may be lower." They suggest that this may be the explanation for the contrathermodynamic syn stereoselectivity exhibited in the addition of many unsymmetrical carbenes to olefins. Although the data for benzenesulfinyl carbene seem to be contrary to their postulations, a closer examination reveals that the stereoselectivity data for sulfenyl, sulfinyl, and sulfonyl substituted carbenes provide a basis for further refinement of their theory. The phenylsulfinyl substituent is large ( A value of 1.929) but the observation of a high degree of anti stereoselectivity is not simply a steric effect, since the sterically larger phenylsulfonyl group ( A value of 2.530) imparts no such selectivity to phenylsulfonyl carbene (anti/syn = 2.1 for addition to c y c l ~ h e x e n e ~ 'Electronically, ), the phenylsulfinyl substituent can stabilize the carbene center by delocalization of its lone pair into the empty orbital on carbon whereas the phenylsulfonyl substituent cannot.32 Increased stability should lead to a more product-like transition state and to a higher degree of anti stereoselectivity. On the other hand, phenylsulfenyl carbene, though stable, is relatively nonstereoselective (anti/syn = 0.77 for addition to cyclohexW e believe that this lack of discrimination is due to the small steric size of the phenylsulfenyl group ( A value of 0.834).Thus, it appears that a substituent must be both stabilizing and large if it is to direct a high degree of anti stereoselectivity. Further studies of diazo sulfoxides and sulfinyl carbenes are underway.

Acknowledgment. W e thank the Robert A . Welch Foundation, Texas Christian University Research Foundation, and the Dow Chemical Company for financial support of this work. References and Notes (1)Part IV of a series, "a-Substituted Sulfoxides": Part Ill, C. G. Venier and H. J. Barager, 111, Org. Prep. Proced. Inf., 6, 77 (1974). (2)W. Kirmse, "Carbene Chemistry", 2nd ed. Academic Press, New York, N.Y., 1971. (3)H. D. Hartzler, J. Am. Chem. SOC.,95,4379 (1973):R . A. Moss and F. G. Pilkiewicz, Synthesis, 209 (1973):D. Seebach, Chem. Ber., 105,487 (1972):R . H. Rynbrandt and F. E. Dutton, Tetrahedron Lett., 1933 (1972);M. Makosza and E. Biatecka, TetrahedronLett., 4517 (1971):A. Krebs and H. Kimling, Angew. Chem., int. Ed. Engl., 10, 509 (1971). (4) W. J. Baron, M. R. Decamp, M. E. Hendrick, M. Jones, Jr., R. H. Levin, and M. B. Sohn, in "Carbenes". Vol. 1, M. Jones, Jr., and R. A. Moss, Ed.. Wiley. New York, N.Y., 1973,Chapter 1 (5)T. Durst. J. Am. Chem. Soc., 91,1034 (1969). (6)a-Diazo sulfoxides have been postulated, but not characterized, twice previou~ly.',~

(7)E. Ayca, Rev. Fac. Sci. Univ. Isfanbui, Ser. C, 22, 371 (1957);Chem. Absfr., 53, 11287/(1959). (8) D. Hodson and G. Holt, J. Chem. SOC.C, 1602 (1968). (9) For a recent attempted preparation of a-amino sulfoxides, see T. Numata and S. Oae, Bull. Chem. Soc. Jpn,, 45, 2794 (1972). (10)T. Kumamoto and T. Mukaiyama, Bull. Chem. SOC. Jpn., 41, 2111 (1968). (11)P. Yates, B. L. Shapiro. N. Yoda. and J. Fugger, J. Am. Chem. Soc.. 79, 5756 (1957). (12)A. M. Van Leusen and J. Strating, R e d Trav. Chim. Pays-Bas, 84, 151 (1965). (13)A. Ledwith and E. C. Friedrich, J. Chem. Soc., 504 (1964). (14)E. Eistert, M. Regitz, G. Heck, and H. Schwall, in "Methoden der Organischen Chemie", 4th ed, E. Muller, Ed., Georg Thieme Verlag. Stuttgart, 1968,pp 473-893.

(15)C. G. Venierand H. J. Barager, 111, J. Chem. SOC.,Chem. Commun., 319 (1973). (16)The four diastereomers of 1,2dimethyl-3-phenylsulfinyl cyclopropane can be analytically separated from each other and from 1 by HPLC on Corasil with 15% chloroform in petroleum ether as the moving phase. Characterization of the 1,2-dimethyl-3-phenylsulfinylcyclopropanesis as follows: (1)cisdimethyl-synphenylsulfinyl (ir(neat), 1080, 1030 cm-' (-S(O)-); NMR (8), 1.1-1.7(complex multiplet dominated by methyl resonances at 1.3 and 1.5,8 H), 2.25 (m. 1 H), 7.5 (m, 5 H); mp 66-68', Anal. Calcd for CI~H~IOS: C. 67.99;H. 7.26.Found: C. 67.94:H, 7.18); (2)cisdimethyl-anfiphenylsulfinyi (ir(neat), 1075, 1035 cm-' (-S(0)-): NMR (8), 1.0(d, 6 H), 1.3-1.8(m, 3 H). 7.5 (m, 5 H): bp. 62-83' (0.01 mm). Anal. Found: C, 67.76;H, 7.46);(3)transdimethyl (a mixture of two diastereomers) (lr(neat), 1080, 1030 cm-l (-S(0)-); NMR (6), 0.81.7 (complex, 8 H), 1.75-2.1(m, 1 H), 7.5 (m, 5 H); bp 83-65' (0.05 mm). Anal. Found: C, 67.89;H, 7.28). 17) R. Helder, T. Doornbos, J. Strating, and B. Zwanenburg, Tetrahedron,

29, 1375 (1973). 18) K. N. Houk, J. Sims. R . E. Duke, Jr.. R. W. Strozier and J. K. George, J. Am. Chem. SOC.,95,7287(1973). 19) K. N. Houk. J. Sims, C. R. Watts, and L. J. Luskus, J. Am. Chem. Soc., 95, 7301 (1973). 20) P. K. Kadaba and T. F. Colturi, J. Heferocycl. Chem., 6,829 (1969). !1) R. Huisoen. H. Stanol. H. J. Sturm. and H. Waaenhofer. Anoew. Chem.. * 73, 17{(1961) 1791 wioaestf the - -, .Preliminarv . - ....... , analvqiq _. _ _ I=__._ that .. .. .- decrease _--_in norcarane - _ - .- vinld ,.-.- results in the formation of bis(phenylsulfiny1)ethene.Decomposition of the diazo compound by adding it to diethyl fumarate or cyclopentane gives this compound also. (23)The syn and anti configurations of the cyclopropyl sulfoxides derived from cyclohexene and 2-2-butene were established by hydrogen peroxide oxidation to the corresponding sulfones, whose configurations had been previously determined.24Characterization of phenylsulfinyl norcaranes is as follows: syn (ir(KBr), 1080, 1030 cm-' (-S(O)-), NMR (6), 1.1-2.5(complex, 1 1 H, JH,-H,= 8.5Hz at ca. 6 2.3),7.3 (m, 5 H); mp 95.5-97'. Anal. Calcd for Ci3H1e0S; C. 70.91;H, 7.27.Found: C, 70.90;H. 7.30.);anti (ir(neat), 1080, 1040 cm-' (-S(0)-), NMR (6). 1.0-2.2(m, 1 1 H. JH,-H~ = 4.5 Hz at 6 2.0),7.3(m, 5 H); bp 132-133' (0.25mm). Anal. Found: C, 70.83;H. 7.50.). (24)U. Schollkopf, G. J. Lehmann, J. Paust, and H.-D. Hartl, Chem. Ber., 97, ~

1527 (1964). (25)Reference 2,pp 286-287. (26)R. A. Moss, in "Carbenes", Vol. 1, M. Jones, Jr., and R. A. Moss, Ed., Wiley, New York, N.Y., 1973,Chapter 2. (27)R. A. Moss in "Selective Organic Transformation", Vol. 1. B. S. Thyagarajan, Ed., Wiley, New York, N.Y.. 1970,pp 35-88. (28)R. Hoffmann, C.C. Levin. and R. A. Moss, J. Am. Chem. Soc., 95,629 (1973). (29)E.L. Eliel, E. W. Della, and M. Rogic, J. Org. Chem., 30, 855 (1965). (30)E. L. Eliel, N. L. Aliinger, S. J. Angyal, and G. A. Morrison, "Conforrnational Analysis", Wiley. New York, N.Y., 1965,p 439. (31)R. A. Abramovitch and J. Roy, Chem. Commun., 542 (1965). (32)That sulfur lone pairs can stabilize carbenes is well documented in the divalent sulfur series.2 For example. (CHzS)2C is so stable that it does not insert into normal (33)U. Schollkopf and E. Wiskott, Angew. Chem., Int. Ed. Engi., 2, 485

(1963). (34)E. L. Eliel and M. H. Gianni, Tefrahedron Left., 97 (1962). (35)Texas Christian University Research Foundation Predoctorai Fellow, 1970-1971;Dow Fellow, 1971-1973;Robert A. Welch Foundation Predoctoral Fellow, 1973-1974. (36)Texas Christian University Undergraduate Scholar, 1972: Dow Fellow. 1973-1 975.

Clifford C.Venier,* Howard J. Barager, 111,35 Mark A. Ward36 Department of Chemistry, Texas Christian University Fort Worth, Texas 76129 Received June 26. I974

A Highly Activated Cyclopropane for Homoconjugate Reactions Sir:

I n contrast to double bonds, which are susceptible to the Michael reaction' by virtue of one activating group (eq 1 , X = C02R, C N , etc), cyclopropanes require two such activating groups for homoconjugate addition. The process formalized in eq 2 for nucleophiles has been reported for enolates,2J-c amines,' mercaptan^,^ enamines,4 and cup rate^.^ However, the severity of the conditions usually required for the intermolecular process even with two activating groups6-* has sharply reduced its attraction as a synthetic meth~d.~-'~ We considered the case of compound 1 in which both activating groups are esters, the most widely studied system in Communications t o the Editor